A few weeks ago I wrote about The Paradox of Choice when ordering food online. Basically, I suggested applying a coin flip algorithm (RNG or Random Number Generator) or setting a default value if you and your friends are stuck choosing food for more than 10 minutes.

Because I am a highly empirical person, last night I actually tested this method while hanging out with three friends at a dorm. The clock already showed half past seven, our stomachs were giving low battery warnings, and the debate over whether to eat satay, spicy noodles, or egg martabak had entered a deadlock phase. Everyone was presenting arguments, but nobody wanted to yield or make a final decision.

I immediately took control. I pulled a coin from my wallet and established a clear rule engine.

If heads, we order satay. If tails, we order egg martabak. The spicy noodles were automatically dropped from the queue because the distance (delivery fee) was too far for our tight end-of-the-month budget. This is called data pre-processing: you filter out outliers before putting the data into the model.

I flipped the coin high into the air. While the coin was spinning, the psychological magic I discussed yesterday actually happened. One of my friends suddenly yelled, "I hope it lands on tails, I am too lazy to eat anything on skewers!"

Bam. Do you see? The human brain actually already knows what it wants. It just needs an external trigger (this coin) to bypass the complicated analysis layer and access its deepest intuition. The coin is not a tool to determine fate, but a tool to test how disappointed you are with one outcome, which ultimately reveals your true preference.

Unfortunately, the coin landed on heads. Satay it is.

According to the SLA (Service Level Agreement) we agreed upon earlier, the algorithm's decision is absolute and cannot be overridden by emotion. My friend who wanted the martabak pouted a bit, but he still agreed because he knew if we debated again, we would only eat at ten PM. This is the essence of satisficing: taking the "good enough" option to bypass a bottleneck.

The order arrived thirty minutes later. The satay was slightly too salty and the chili sauce lacked a kick. My friend who lost the coin bet started making passive-aggressive jokes. But honestly, the imperfect taste of this satay was far better than starving while ruining our friendship over different tastes.

This experiment proved one important thing: in a real world full of probability and uncertainty, the execution time of making a decision is often more crucial than the quality of the decision itself. You could spend an hour finding the best satay in the city, but by the time the satay arrives, you have already caught a cold and lost your appetite.

Sometimes, being a programmer in real life means you must be brave enough to delegate your mental load to a random system, trust the run results, and accept a small error rate to maintain the survival of your main server.

Khay